基于嵌入训练阵列的OTFS信道估计

Training Embedded Array-Based OTFS Channel Estimation

正交时频空(Orthogonal Time Frequency Space,OTFS)调制在快时变高多普勒信道中展现出良好的频谱利用率和显著的误码率性能优势。然而,传统的OTFS信道估计方法频谱效率较低,且无法应用于低轨卫星等具有高多普勒频偏的通信场景中。文章提出了基于嵌入训练(Training Embedded,TE)阵列的信道估计方法,利用训练阵列的完美自相关特性进行准确的信道估计,并提高了频谱效率。同时,针对低轨卫星通信高多普勒频偏的特性,本文利用快速傅里叶变换对路径的整数倍频偏进行估计,补偿并消除列相关相位偏移干扰,提升了低轨卫星通信场景下信道估计的准确度。仿真结果表明,所提方法在地面、低轨通信场景中均具有较高信道估计精度和频谱效率。

Orthogonal time frequency space modulation(OTFS)shows great spectrum efficiency and significant bit error rate performance in fast time-varying channels with high Doppler.However,the traditional OTFS channel estimation method suffers from low spectrum efficiency and is unable to be applied in communication scenarios with high Doppler such as low earth orbit(LEO)satellite communication.In this paper,we propose an OTFS frame structure with the training embedded array.Utilizing the perfect autocorrelation of the training embedded array,accurate channel estimation is performed,and the spectrum efficiency is improved.Furthermore,according to the characteristics of high Doppler in LEO satellite communication,we use the fast Fourier transform to estimate the integer multiple Doppler shifts for each path and eliminate the interference of column correlation phase offset.In this way,channel estimation accuracy in low-orbit satellite communication.Simulation results show that the proposed method has high channel estimation accuracy and great spectral efficiency on both ground and LEO satellite communication.